In this work, we experimentally study the emergent behavior of a system of agents that make decisions based on the states of their neighbors. The decision model is stochastic and accounts for a limited number of neighbors within a defined communication radius.

We use Kilobots for the experiments: commercial robots capable of sending and receiving messages, moving, and processing information. The experiments are conducted with 20 robots placed inside a circular arena, where each robot can be in one of two states of motion: rotating clockwise or counterclockwise. Each robot transmits its state to its neighbors and receives the states of those around it.

The collective behavior of the system is characterized by measuring the global rotation direction as a function of the number of neighbors considered in the decision-making process and the stochastic parameter of the model. Subsequently, an anonymous leader is introduced into the system by programming one of the agents with a deterministic periodic behavior. The global rotation of the system is measured again, this time as a function of the leader's period and the stochastic decision parameter of the other agents.

The results of this study show how an anonymous leader can influence the collective decision-making of the system.